#continuousrefinementbased 搜尋結果
A THREAD FOR YOU ALL 🧵📌- CONTINUATION SETUP📊 It is not always about disrespected A-shape or V-shape keylevel, there are more context to it than you know💁 But before we continue, please like, repost, share to your socials💻📱and stick with me till the end Let's move on🚶🚶
As part of our recent work on memory layer architectures, I wrote up some of my thoughts on the continual learning problem broadly: Blog post: jessylin.com/2025/10/20/con… Some of the exposition goes beyond mem layers, so I thought it'd be useful to highlight separately:
📌 What Is Continual Improvement? 📈 Continual improvement is the heartbeat of ISO systems. It’s not about perfection, it’s about progress. 🔍 Ways to drive continual improvement: - Regular feedback loops - Lessons learned from incidents - Small process tweaks that add up 📌…
We’re excited to share our new paper: Continuously-Augmented Discrete Diffusion (CADD) — a simple yet effective way to bridge discrete and continuous diffusion models on discrete data, such as language modeling. [1/n] Paper: arxiv.org/abs/2510.01329
![UnderGroundJeg's tweet image. We’re excited to share our new paper: Continuously-Augmented Discrete Diffusion (CADD) — a simple yet effective way to bridge discrete and continuous diffusion models on discrete data, such as language modeling. [1/n]
Paper: arxiv.org/abs/2510.01329](https://pbs.twimg.com/media/G2jrCUDaMAAojey.png)
Base has been one of the few true constants since I joined CT — a north star for reliability and execution.
Happy to share code and videos of my paper Continuous Remeshing for Inverse Rendering😀 github.com/Profactor/cont…
📊 Consistency alone is overrated. It doesn’t produce lasting results for you. This is what actually work!!!! Here’s why people remain consistent yet see little or no growth.
🚫Repetition isn’t progress. 🌀Familiar doesn’t mean effective. 🏛At CQCapitalbio Inc, we don’t act to stay busy, we act to shift direction. 🎯Progress means choosing better, not doing more. #CQCapitalbio #BeyondRepetition #SmartDecisions #CapitalProgress #DirectionalThinking
You can’t build AI products like other products, for two reasons: 1. AI products are inherently non-deterministic 2. Every AI product must negotiate a tradeoff between agency and control When companies don’t recognize these differences, their AI products face unexpected…
⚠️ Failure may surprise. Resilience shouldn’t. 🧠 At CQCapitalbio Inc, redundancy is embedded not as backup, but as core design. Every mirrored logic and alternate route keeps the system steady when shocks hit. 🛡️ Resilience by design. Stability by intent. #CQCapitalbio
Speaking now: Shaik Buden Saheb The hidden layer of observability - Continuous Profiling for peak performance! 📈🔬 #Observability #ContinuousProfiling #CNCFHyderabad
continue - stay based 🟦 start - more incentive programs for builders, scaled and automated like farcaster weekly dev rewards beside well-designed initiatives like base batch stop - not stop 🫡
Our mission is to preserve Ethereum's core value: composability. Based Espresso is our latest development towards a future of composable L2s. This design sees L2s retain autonomy over their sequencing rights, and supports any Ethereum L2, from based rollups to validiums:
🚀 Continue CLI is here! The async coding agent that actually understands your codebase. Making AI continuous in your dev workflow. - Stream AI responses in real-time - Run parallel background tasks - Smart commit messages, code analysis & more
With Continue CLI + Agents, you don’t have to pause coding to “explain” context to an external tool. Run structured, repeatable AI actions from the terminal or in the background. Refactors, tests, docs, all without leaving your flow. No more cognitive reset tax. Read more 👇
Can LLMs enhance their own output without human guidance? In some cases, yes! With Self-Refine, LLMs generate feedback on their work, use it to improve the output, and repeat this process. Self-Refine improves GPT-3.5/4 outputs for a wide range of tasks. selfrefine.info
Something I'm learning is that dopamine can come from refinement, not novelty. Chasing subtle progress builds momentum through each micro-adjustment. Every time something doesn't work, I figure out how to fine-tune it so that it does.
Something went wrong.
Something went wrong.
United States Trends
- 1. #WWERaw 12K posts
- 2. Cuomo 118K posts
- 3. Koa Peat N/A
- 4. Cowboys 30.2K posts
- 5. #OlandriaxCFDAAwards 6,985 posts
- 6. Monday Night Football 8,555 posts
- 7. Walt Weiss 2,746 posts
- 8. Caleb Wilson N/A
- 9. Harvey Weinstein 8,511 posts
- 10. Teen Vogue 4,996 posts
- 11. Braves 11.7K posts
- 12. Arizona 30.6K posts
- 13. Sliwa 33.3K posts
- 14. Schwab 5,859 posts
- 15. #RawOnNetflix N/A
- 16. Diane Ladd 6,643 posts
- 17. Hamburger Helper 2,826 posts
- 18. Ben Shapiro 40.2K posts
- 19. Myles Turner 1,671 posts
- 20. Stefan Vaaks N/A